1. Field of the Invention
The present invention concerns the field of dendritic polymers where dendrimers are an example of the preferred polymers. These polymers have void spaces that may entrap molecules and their surface functionalities may undergo further reactions.
2. Description of Related Art
Branched Polymer Ring Opening Reactions
Various ring opening reactions to prepare branched polymer systems are known. A few of these processes are described below.
Polymerizations using ring opening is well known, particularly with using cyclic ethers, amides, aziridenes, sulfides, siloxanes and others by either anionic, cationic or other mechanisms. (George Odian, Principles of Polymerization, pub. John Wiley and Sons, 1993, Chapter 7.) However, use of ring opening polymerizations in the synthesis of highly branched polymers is less well known. Work has been done in the use of ring opening polymerizations in the synthesis of various hyperbranched polymers. In most of the cases the ring opening polymerization is of the traditional type resulting in random hyperbranched polymers with broad polydispersities.
One of the first examples of ring opening polymerizations to prepare a hyperbranched polymer was the work of Odian and Tomalia [P. A. Gunatillake, G. Odian, D. A. Tomalia, Macromolecules, 21, 1556 (1998)] where hyperbranch materials were made from oxazolines.
Ring opening has been used in the generation of linear or comb-branched polyethers as single ion conductors [X. G. Sun, J. B. Kerr, C. L. Reeder, G. Liu, Y. Han, Macromolecules, 37(14), 5133-5135 (2004)].
Ring-opening polymerization of 2-hydroxymethyloxetane under basic conditions was attempted to obtain hyperbranched polyethers [Y. H. Kim, J. Polym. Sci., Polym. Chem., 36, 1685 (1998)].
D. A. Tomalia's work on ring opening polymerization of oxazolines achieved hyperbranched PEOX or PEI polymers (See U.S. Pat. Nos. 4,690,985, 5,631,329, and 5,773,527).
Hyperbranched dendritic macromolecules have been made using a multibranching polymerization (“MBP”) approach with an initiator at the core, involving ring-opening polymerization including, for example, Pd-catalyzed ring opening polymerization of cyclic carbamates in the presence of an initiator using oxazinones [M. Suzuki; A. Ii, T. Saegusa, Macromolecules, 25, 7071-7072 (1992), and M. Suzuki, S. Yoshida; K. Shiraga, T. Saegusa, Macromolecules, 31, 1716-19 (1998)].
Epoxide ring opening, involving an AB2 type monomer polymerization, is initiated by addition of a catalytic amount of an initiator, such as hydroxide ion, and goes through a novel propagation mode distinct from other hyperbranched polymer methods involving acid- or base-catalyzed reactions [H. T. Chang, J. M. J. Frechet, J. Am. Chem. Soc., 121, 2313-2314 (1999)]. AB2 monomer type glycidols are polymerized to hyperbranched “polyglycerols” by controlled anionic ring opening polymerization to polydispersities below 1.5 [A. Sunder, R. Hanselmalm, H. Frey, R. Mulhaupt, Macromolecules, 32, 4240-4246 (1999)]. Cationic cyclopolymerization of dianhydro-D-mannitol is used to produce hyperbranched carbohydrate polymer [T. Imai, T. Satoh, H. Kaga, N. Kaneko, T. Kakuchi, Macromolecules, 36, 6359-6363 (2003); T. Imai, T. Satoh, H. Kaga, N. Kaneko, T. Kakuchi, Macromolecules, 37, 3113-3119 (2004)].
Hyperbranched polymers are obtained by combining ring-opening polymerization and some features of self condensing vinyl polymerization (“SCVP”), ring opening polymerization of caprolactone to give hyperbranched polyesters having a polydispersity of about 3.2 [M. Liu, N. Vladimirov, J. M. J. Frechet, Macromolecules, 32, 6881-6884 (1999)].
Ring opening polymerization of bis(hydroxymethyl)caprolactones gave hyperbranched polyesters [M. Trollsas, P. Lowenhiehm, V. Y. Lee, M. Moller, R. D. Miller, J. L. Hedrick, Macromolecules, 32, 9062-9066 (1999)].
Cationic ring opening polymerization of ethyl hydroxymethyl oxetanes resulted in hyperbranched polyethers, polydispersities in the range of 1.33-1.61 [Y. Mai, Y. Zhou, D. Yan, H. Lu, Macromolecules, 36, 9667-9669 (2003)].
3-Ethyl-3-(hydroxymethyl)oxetane ring opening is used to generate hyperbranched polyethers [H. Magnusson, E. Malmstrom, A. Hult, Macromolecules, 34, 5786-5791 (2001)].
Dendritic polypeptides were obtained by ring opening polymerization of N-carboxyanhydrides. The method involves repetitive sequences of N-carboxyanhydride ring opening and end-coupling reactions. This process results in polymeric regions with a statistically driven average chain length per branch, having no precise lengths, and results in a polymer with typical polydispersities of 1.2-1.5.
Precise Dendrimer Ring Opening Reactions
Polysulfide dendrimers can be formed by reacting a polythiol under basic conditions with epichlorosulfide to form polyepisulfides (See U.S. Pat. Nos. 4,558,120, and 4,587,329). These same patents also discuss the preparation of a polyaminosulfide dendrimer using a reaction of a polyamino core with an excess of ethylene sulfide to form a polysulfide followed by reaction with excess aziridine to from further generations.
Addition of N-tosyl aziridine is discussed as a way to create a partially protected dendrimer surface (U.S. Pat. Nos. 4,361,337; 4,587,329; and 4,568,737) and is extended to azetidine derivatives.
Precise Dendrimer Ring Opening Reactions for Attachment of Surface Groups
Ring opening reactions are discussed as a way to add terminal groups. For example, U.S. Pat. No. 4,568,737 discloses the use of oxiranes to create a polyol surface on the dendrimer.
Processes for Precise Dendrimer Structures
Many specific reactions have been used to create a wide range of precise dendrimer structures. These reactions typically define a core (“C”), branch structure type (“BR”) and terminal functionality (“TF”). The synthesis of precise dendrimer structures has been performed using two broad approaches that have been categorized as “convergent synthesis” and “divergent synthesis” [Dendrimers and other Dendritic Polymers, eds. J. M. J. Frechet, D. A. Tomalia, pub. John Wiley and Sons, (2001)]. Within these broad categories there are further variations regarding branch cell construction (i.e., in-situ and preformed) or dendron anchoring type construction.
One of the earliest published uses of branch cell reagents involved coupling preformed branch cells around a core to form low molecular weight arborol structures [G. R. Newkome, Z.-Q. Yao, G. R. Baker, V. K. Gupta, J. Org. Chem., 50, 2003 (1985)]. Poly(thioether) dendrimers were synthesized using protected, preformed branch cell reagents based on a pentaerythritol core; Nc=4 and 4-acetothiomethyl-2,6,7-trioxabicyclo[2.2.2]octane; Nb=3. In this case a protected branch cell reagent was used in the building of the dendrimer branch structure, which requires chemical deprotection as an added step to rapidly build structure. Although the reagent used is a polycyclic type ether (i.e., orthoester), the ether ring is not strained and does not ring open during polymerization.
Steric Effects in Traditional Small Molecule Chemistry
Steric effects, as defined in small molecule chemistry, are due to the volume of sub-nanoscale space (i.e., 0.05-1 nm) that all fundamental small molecule “building block components” (i.e. atoms, functional groups, hydrocarbon scaffolding, etc.) occupy and their relationship to each other in critical reaction and assembly events. The effect that their relative sizes have on reactivity, displacements, substitutions, chirality, associations, assemblies, specific product formation and attainable architectures have always remained issues of very high importance both in the academic as well as commercial realms. For example the steric effect that decreases reactivity is call “steric hinderance” [See P. Y. Bruice, Organic Chemistry, 2nd Ed. (1998), p 362, Prentice Hall]. Steric hinderance results from groups getting in the way at a reaction site. Classical examples include the “neopentyl effect”, wherein the relative reactivities of increasingly hindered alkyl halides to SN2 reactions are increasingly suppressed to a point that that a teriary alkyl halide (i.e. neopentyl bromide) is too slow to measure. It is not just the number of alkyl groups attached to the carbon undergoing nucleophilic attack that determines the reaction rate; the relative sizes of the alkyl groups are also very important. Cram's Rule_is another classical example of a small molecule steric effect. While not wishing to be bound by theory, it is believed that steric effects control the stereo selective reactivity at a carbonyl oxygen resulting in chiral introduction. Cram's Rule states that a nucleophile approaches a carbonyl along the smallest substituent alignment. The largest group aligns itself anti to the carbonyl group to minimize the steric effect such that the nucleophile preferentially attacks from the side of the small substituent. [See D. J. Cram, A. Elhafez, J. Am. Chem. Soc. 74, 5828 (1952).]
These above brief examples not only portend the possibility but also the importance that such analogous “steric effects” may offer if discovered and defined for critical construction components at the nanoscale level, (i.e. 1-100 nm). The nanoscale rules for these NSIS effects are virtually unknown. How NSIS relates to this invention is described in the Detailed Description of this specification.
Poly(amidoamine) Dendrimer (“PAMAM”) Synthesis
Some of the difficulties in the synthesis of dendrimers are inherent in the methods used to make them. For example the preparation of poly(amidoamine) (“PAMAM”) dendrimers, one of the key compositional families of these dendritic polymers, currently focuses on Michael addition chemistry with in-situ branch cell formation [Dendrimers and other Dendritic Polymers, eds. J. M. J. Frechet, D. A. Tomalia, pub. John Wiley and Sons, (2001), Chapter 25]. The usual process includes an amidation step which involves slow chemistry, long reaction times and non-differentiated difunctional intermediates. These circumstances force the process to require high dilutions resulting in low capacities and high costs, particularly at higher generations. Additionally, PAMAM dendrimers, due to their specific amide structures have access to low energy routes to degradation through reverse Michael addition reactions and hydrolysis reactions.
Clearly, it would be desirable to find a process to make precise dendrimer structures with a faster reaction time, easier separation with fewer by-products, and lower cost of manufacture than that presently used. Additionally, if the dendrimers were more stable and easier to scale, that also would be desired.